Author |
|
jeffw_00 Super User
Joined: June 30 2007
Online Status: Offline Posts: 929
|
Posted: September 19 2021 at 16:47 | IP Logged
|
|
|
Hi Dave - I am seeing repeatable communication issues with the
new release that never happened with the old one. I'm wondering if
some timeout or other parameters have changed. I've uploaded a
log excerpt and marked the worrisome sections. These are issues
that don't occur with the old release (ever) and have repeated 3 or
4 times in the last couple of hours since I switched over. (i've since
switched back). Any/all advice appreciated. I'm using a Serial PLM
if that helps.
Thanks!
/j
2021-
09-19_164718_commprob.txt
Edited by jeffw_00 - September 19 2021 at 17:17
|
Back to Top |
|
|
jeffw_00 Super User
Joined: June 30 2007
Online Status: Offline Posts: 929
|
Posted: September 19 2021 at 16:52 | IP Logged
|
|
|
related question - I have a 2 powerhome folders on my C drive - one
with 2.2 and one with 2.1.5. Since my C drive is an SSD I also have 2
such folders on my D (regular HDD) drive that hold powerhome.db.
Right now, when I swap versions I swap both folders (via rename).
But I think that when I swap from 2.1.5 to 2.2, I could also (after
copying it) leave the old D drive folder with the 2.1.5 database init and
just run PHUPG again before firing up 2.2 - true?
Thanks!
/j
|
Back to Top |
|
|
dhoward Admin Group
Joined: June 29 2001 Location: United States
Online Status: Offline Posts: 4447
|
Posted: September 20 2021 at 23:43 | IP Logged
|
|
|
Jeff,
Concerning your log...the line below is a problem:
Communication problem (No Response) with Insteon controller PLMMODEM detected. Attempting to restart controller (1 of 5)...
I'll need to trace the code to see what causes this problem. The following errors associated with the macro are probably the result of the
insteon controller not being fully restarted. I'll probably need to make some changes where a controller restart doesnt go through the database
discovery again (like it does on startup and required a delay in your macro). I'll probably need a flag as well so that functions return a
specific value when a command to the controller is sent when the controller is not fully started.
The next highlighted block with the line:
Incoming Insteon received on INSTEON PLM controller PLMMODEM. From Address:[48.6C.A8] To Address:[13.00.01] Flags:[199] Cmd1:[6] Cmd2:[0].
Decode: Group Broadcast Success/Failure Report.
This is not an error. This is just logging a new type of error message that Insteon sends. In the previous version of PowerHome, you would have
gotten a log entry but it would have said something like unknown command. This command is sent when an Insteon device is manually controlled
(that is linked to other Insteon devices). The Insteon device first sends out a grYou are oup broadcast followed by group cleanup commands.
After the group cleanup commands the group broadcast success/failure command is sent. In this particular line, the GARAGESW was manually
controlled, it sent the group broadcast followed by group cleanups. A group cleanup is an acknowledged command and if a device is not actually
acknowledged, this command will report that. So not an error or anything but a normal Insteon command that comes in that is logged with the
correct description vs the previous unknown command nomenclature.
******************************
Concerning your question of versions...If the new version is installed in c:\powerhome and the previous version is in something like
c:\powerhome_215, you wouldnt necessarily need to rename folders to switch between versions. Just update the appropriate pwrhome.ini to point to
the proper database location. So the pwrhome.ini file c:\powerhome_215 might have the database at d:\powerhome_215\database. Unless you're
specifically doing something in your PowerHome code that is specifying a directory path, this should work for you to quickly switch back and
forth.
You are correct concerning the database. If you've switched back to 2.1.5 and that database is more current and you want to try 2.2 again, you
can just copy your 2.1.5 database to the 2.2 directory and run the PHUPG on it so your 2.2 testing will have the latest data. Just be careful
you don't update your only copy of 2.1.5 as you cant go backwards.
I'll let you know what I find out concerning the controller restart.
Dave.
|
Back to Top |
|
|
jeffw_00 Super User
Joined: June 30 2007
Online Status: Offline Posts: 929
|
Posted: September 21 2021 at 07:42 | IP Logged
|
|
|
Hi Dave - thanks so much for getting back to me and for the
thorough explanations and answers. I think you're saying that the
problem I'm seeing is that some sort of "Communication Problem"
is triggering a restart, and in 2.2 (unlike 2.1.5) that restart
"distracts" (for lack of a better word) PH so that some subsequent
commands don't happen properly. What I wonder (and I can look if
you tell me what to look for) is whether I also get the
"Communication Problem" with the older code but it just causes no
adverse effect, or whether I get the Communication Problem with
only the newer code. If the latter, I hope your proposed fix will be
sufficient, but wonder whether I need to root cause why I'm getting
the Communication Problem at all. One theory I had is that
perhaps I'm an outlier in that I use a PLM with a serial (as opposed
to USB) interface? Anyway, -many- thanks for looking into this.
Let me know what else I can do to help. Standing by :-)
/j
|
Back to Top |
|
|
dhoward Admin Group
Joined: June 29 2001 Location: United States
Online Status: Offline Posts: 4447
|
Posted: September 22 2021 at 23:17 | IP Logged
|
|
|
Jeff,
Yes, your summation is correct. Im pretty sure the previous version of PowerHome had the communication problem detection but it would
have done a restart and then been done almost instantly. The current version rescans the whole PLM database is which is the cause of
your current problem. You can search the log to see if you had any restarts with 2.1.5 and they should show but it's possible that the
newer code in 2.2 is using different criteria to signify a communication problem. I will need to review that whole section of code to
know for sure.
Serial vs USB shouldnt make a difference. Actually, I prefer the straight serial as they are definitely more stable. The USB and
Serial PLM's use the same code and are communicated with as a serial port with the USB creating a virtual serial port that PowerHome
connects to so an extra layer of potential problems with the USB device.
I'll let you know what I find out after Ive been able to go through the code.
Dave.
|
Back to Top |
|
|
jeffw_00 Super User
Joined: June 30 2007
Online Status: Offline Posts: 929
|
Posted: September 23 2021 at 07:47 | IP Logged
|
|
|
Hi Dave - FWIW - I searched my 2.1.5 log for (case-insensitive)
"communication problem" and "no response" and found nothing in the
past 2 weeks (excluding the short periods I was running .2.2 - less
than several hours). So yeah - something in 2.2 is detecting an issue
where 2.1.5 doesn't.
I've been on 2.1.5 this long, I can stay on it longer :-). Thanks in
advance for anything you can do,
Thanks!
/j
|
Back to Top |
|
|
dhoward Admin Group
Joined: June 29 2001 Location: United States
Online Status: Offline Posts: 4447
|
Posted: September 24 2021 at 21:44 | IP Logged
|
|
|
Jeff,
Ive added it to my notes to look into this to see what might be different between 2.1.5 and 2.2.
Dave.
|
Back to Top |
|
|
jeffw_00 Super User
Joined: June 30 2007
Online Status: Offline Posts: 929
|
Posted: September 25 2021 at 07:46 | IP Logged
|
|
|
No problem. . Patiently standing by
|
Back to Top |
|
|
jeffw_00 Super User
Joined: June 30 2007
Online Status: Offline Posts: 929
|
Posted: November 21 2021 at 09:26 | IP Logged
|
|
|
H Dave - any update here? Thanks!
|
Back to Top |
|
|
jeffw_00 Super User
Joined: June 30 2007
Online Status: Offline Posts: 929
|
Posted: March 29 2022 at 08:02 | IP Logged
|
|
|
Hi Dave - any update here? I'd really like to upgrade at some point :-)
|
Back to Top |
|
|
jeffw_00 Super User
Joined: June 30 2007
Online Status: Offline Posts: 929
|
Posted: April 10 2022 at 13:16 | IP Logged
|
|
|
Perhaps you can add an option to disable the rescan on restart.
|
Back to Top |
|
|
BeachBum Super User
Joined: April 11 2007 Location: United States
Online Status: Offline Posts: 1880
|
Posted: July 03 2022 at 22:38 | IP Logged
|
|
|
I too am having more down time than up time due to many
communications errors. It can easily be duplicated by
plugging in an invalid comport number. PH locks itself
into an endless nightmare of error recovery. I don't have
bad memory as in Kens nightmare. I'm going back and see
what I did wrong, or Dave, to something of stability.
After that I'll investigate what happen and gather more
documentation. The problem as I see it is that PH loses
its way and gets mired down in muddles of errors. It is
in general as not as specify to a particular problem.
Sorry Jeff didn't mean to hijack your thread. Just
thought it would be simpler to add my comments
rather than reinvent the wheel.
Pete...
__________________ Pete - X10 Oldie
|
Back to Top |
|
|
jeffw_00 Super User
Joined: June 30 2007
Online Status: Offline Posts: 929
|
Posted: July 04 2022 at 07:37 | IP Logged
|
|
|
No problem Beachbum. With the latest patch I'm seeing the
messages rarely - and it's not impacting operation. It seems to
happen when I have a macro that has insteon commands separated
by hundreds of milliseconds (i.e. using DELAY). If you don't have the
latest patch you should try it.
|
Back to Top |
|
|